Envit5-tuned / README.md
gozu888's picture
Update README.md
2ac5a9c
|
raw
history blame
1.13 kB
metadata
license: apache-2.0
datasets:
  - mt_eng_vietnamese
language:
  - en
  - vi
tags:
  - translation
  - t5
  - en-vi
  - vi-en

gozu888/Envit5-tuned

This model is a fine-tuned version of VietAI/envit5-translation on the mt_eng_vietnamese dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1988

Model Details

Model Description

Hyperparameters

The following hyperparameters were used during training:

    model.train(
        train_df=train_df,
        eval_df=test_df,
        source_max_token_len=128,
        target_max_token_len=128,
        batch_size=32,
        max_epochs=15,
        use_gpu=True,
        save_only_last_epoch=True,
        precision=16,
        dataloader_num_workers=4
    )